Related Experiment Video
Updated: Aug 2, 2025
Detection and Isolation of Cancer in Prostate Biopsies Using Stimulated Raman Histology and Artificial Intelligence
Published on: June 10, 2025
A comparative study of the inter-observer variability on Gleason grading against Deep Learning-based approaches for
José M Marrón-Esquivel1, L Duran-Lopez2, A Linares-Barranco2
1Robotics and Tech. of Computers Lab., Universidad de Sevilla, 41012 Seville, Spain; Escuela Técnica Superior de Ingeniería Informática (ETSII), Avenida de Reina Mercedes s/n, Universidad de Sevilla, 41012 Seville, Spain; Escuela Politécnica Superior (EPS), Universidad de Sevilla, 41011 Seville, Spain.
Background:
Among all the cancers known today, prostate cancer is one of the most commonly diagnosed in men. With modern advances in medicine, its mortality has been considerably reduced. However, it is still a leading type of cancer in terms of deaths. The diagnosis of prostate cancer is mainly conducted by biopsy test. From this test, Whole Slide Images are obtained, from which pathologists diagnose the cancer according to the Gleason scale. Within this scale from 1 to 5, grade 3 and above is considered malignant tissue. Several studies have shown an inter-observer discrepancy between pathologists in assigning the value of the Gleason scale. Due to the recent advances in artificial intelligence, its application to the computational pathology field with the aim of supporting and providing a second opinion to the professional is of great interest.
Method:
In this work, the inter-observer variability of a local dataset of 80 whole-slide images annotated by a team of 5 pathologists from the same group was analyzed at both area and label level. Four approaches were followed to train six different Convolutional Neural Network architectures, which were evaluated on the same dataset on which the inter-observer variability was analyzed.
Results:
An inter-observer variability of 0.6946 κ was obtained, with 46% discrepancy in terms of area size of the annotations performed by the pathologists. The best trained models achieved 0.826±0.014κ on the test set when trained with data from the same source.
Conclusions:
The obtained results show that deep learning-based automatic diagnosis systems could help reduce the widely-known inter-observer variability that is present among pathologists and support them in their decision, serving as a second opinion or as a triage tool for medical centers.
